We had a file system that recently went from 50% to 100% of available inodes consumed over the period of a couple of months.
Check_MK didn’t generate any alerts regarding inode consumption and reported status OK during the entire timeframe. I can see the historical data in the graphs that it accurately tracked the consumption of
inodes over time, just didn’t alert.
Is there any way to get the check_mk-df check to alert for inode consumptions?
Which version are you using? I recall in 1.2.4 ish I wrote my own plugin for inode usage. I am now on 1.2.8 with WATO and the filesystem check supports inode monitoring and thresholds.
Paul
···
On Mon, Mar 20, 2017 at 7:49 AM, Hanby, Mike mhanby@uab.edu wrote:
Howdy,
We had a file system that recently went from 50% to 100% of available inodes consumed over the period of a couple of months.
Check_MK didn’t generate any alerts regarding inode consumption and reported status OK during the entire timeframe. I can see the historical data in the graphs that it accurately tracked the consumption of
inodes over time, just didn’t alert.
Is there any way to get the check_mk-df check to alert for inode consumptions?
Did you specify thresholds for inodes already?
There are no default thresholds for inode usage, so you will have to
create a rule for this.
Regards,
Evy
···
On Mon, Mar 20, 2017 at 02:49:34PM +0000, Hanby, Mike wrote:
Howdy,
We had a file system that recently went from 50% to 100% of available inodes
consumed over the period of a couple of months.
Check_MK didn't generate any alerts regarding inode consumption and reported
status OK during the entire timeframe. I can see the historical data in the
graphs that it accurately tracked the consumption of inodes over time, just
didn't alert.
Is there any way to get the check_mk-df check to alert for inode consumptions?
Which version are you using? I recall in 1.2.4 ish I wrote my own plugin for inode usage. I am now on 1.2.8 with WATO and the filesystem check supports inode monitoring and thresholds.
Paul
On Mon, Mar 20, 2017 at 7:49 AM, Hanby, Mike mhanby@uab.edu wrote:
Howdy,
We had a file system that recently went from 50% to 100% of available inodes consumed over the period of a couple of months.
Check_MK didn’t generate any alerts regarding inode consumption and reported status OK during the entire timeframe. I can see the historical data in the graphs that it accurately tracked the consumption of
inodes over time, just didn’t alert.
Is there any way to get the check_mk-df check to alert for inode consumptions?